Accuracy of complete-arch dental impressions: a new method of … · 2020-03-19 · in CAD/CAM...
Transcript of Accuracy of complete-arch dental impressions: a new method of … · 2020-03-19 · in CAD/CAM...
Zurich Open Repository andArchiveUniversity of ZurichMain LibraryStrickhofstrasse 39CH-8057 Zurichwww.zora.uzh.ch
Year: 2013
Accuracy of complete-arch dental impressions: a new method of measuringtrueness and precision
Ender, Andreas ; Mehl, Albert
Abstract: STATEMENT OF PROBLEM: A new approach to both 3-dimensional (3D) trueness andprecision is necessary to assess the accuracy of intraoral digital impressions and compare them to con-ventionally acquired impressions. PURPOSE: The purpose of this in vitro study was to evaluate whethera new reference scanner is capable of measuring conventional and digital intraoral complete-arch impres-sions for 3D accuracy. MATERIAL AND METHODS: A steel reference dentate model was fabricatedand measured with a reference scanner (digital reference model). Conventional impressions were madefrom the reference model, poured with Type IV dental stone, scanned with the reference scanner, andexported as digital models. Additionally, digital impressions of the reference model were made and thedigital models were exported. Precision was measured by superimposing the digital models within eachgroup. Superimposing the digital models on the digital reference model assessed the trueness of eachimpression method. Statistical significance was assessed with an independent sample t test (�=.05). RE-SULTS: The reference scanner delivered high accuracy over the entire dental arch with a precision of 1.6±0.6 µm and a trueness of 5.3 ±1.1 µm. Conventional impressions showed significantly higher precision(12.5 ±2.5 µm) and trueness values (20.4 ±2.2 µm) with small deviations in the second molar region(P<.001). Digital impressions were significantly less accurate with a precision of 32.4 ±9.6 µm and atrueness of 58.6 ±15.8µm (P<.001). More systematic deviations of the digital models were visible acrossthe entire dental arch. CONCLUSIONS: The new reference scanner is capable of measuring the precisionand trueness of both digital and conventional complete-arch impressions. The digital impression is lessaccurate and shows a different pattern of deviation than the conventional impression.
DOI: https://doi.org/10.1016/S0022-3913(13)60028-1
Posted at the Zurich Open Repository and Archive, University of ZurichZORA URL: https://doi.org/10.5167/uzh-88978Journal ArticleAccepted Version
Originally published at:Ender, Andreas; Mehl, Albert (2013). Accuracy of complete-arch dental impressions: a new method ofmeasuring trueness and precision. Journal of Prosthetic Dentistry, 109(2):121-128.DOI: https://doi.org/10.1016/S0022-3913(13)60028-1
Dental full arch impression, a new high accurate method of measuring trueness
and precision
A. Ender, A. Mehl
Clinic of Preventive Dentistry, Periodontology and Cariology; Division of Computerized Restor-ative Dentistry, University of Zurich, Zurich, Switzerland
Abstract
Objectives: With the digital intraoral impression a new class of impression techniques is
introduced. To assess the accuracy of these impressions and compare it to conventio-
nal intraoral impression technique threedimensional trueness and precision measure-
ments are necessary. The aim of this study was to evaluate wether a new reference
scanner is capable to measure these threedimensional deviations occuring with conven-
tional and digital intraoral impression techniques of full arch impressions of an in vitro
model.
Methods: A steel reference model was scanned with the reference scanner to evaluate
precision and trueness. The reference model then was used to perform five conventio-
nal impressions with a polyvinylsiloxanether material (Identium, Kettenbach) in a putty
and wash technique with standard stock trays (ASA Permalock, ASA Dental). The con-
ventional impressions were poured with Type IV stone (CamBase, Dentona) and scan-
ned with the reference scanner. Five digital impressions with a optical intraoral scanning
system (CEREC AC, Sirona) were made. In each group, the models were superimposed
and the differences computed with a signed nearest neighbour method. The 90-10%/2
percentile of the differences from each comparison was taken to compute the mean va-
lue for precision. The trueness of each impression method was assessed through su-
perimposition of the impressions with the refefence scan of the steel reference model.
Results: The reference scanner delivers an accuracy with 1.6±0.6 µm for precision and
5.3±1.1 µm for trueness over a full dental arch scan. The conventional impression me-
thod shows significant higher (p<0.001) precision (12.5±2.5 µm) and trueness (20.4±2.2
µm) with only little amount of larger deviations at the second molar region. The digital
impression method was significantly less accurate (p<0.001) with a precision of
32.4±9.6 µm and trueness of 58.6±15.8 µm. More systematic deviation of the digital
model is visible across the entire dental arch.
Significance:The new reference scanner is capable to measure precision and trueness
of both conventional and digital full arch impressions. The digital intraoral impression is
less accurate and shows a complete different pattern of deviation than the conventional
impression method.
Introduction
Dental impressions are a major step in restorative dentistry. They transfer the intraoral
situation of the preparation, the adjacent teeth, the soft tissue and the antagonist to an
extraoral model. This model is then used to produce the final restoration.
The accuracy of this model influences the fit or the restorations which is a major factor
for longevitiy of the restoration (Wettstein, Sailer, Roos, & Hammerle, 2008)(Perakis,
Belser, & Magne, 2004)(Persson, ODEN, ANDERSSON, & Sandborgh-Englund, 2009).
Todays gold standard impression technique ist the physical impression with elastomeric
impression material and stock trays (CI). This negative form of the patients teeth is then
poured with stone, resulting in a physical gypsum model. Various techniques are
described in literature in order to archieve most accurate results (Piwowarczyk, Ottl,
Buchler, Lauer, & Hoffmann, 2002)(Chandran, D. Jagger, R. Jagger, & Barbour, 2010).
With the development of CAD/CAM systems and especially the use of zirkoniumdioxid
for dental restorations, the digital model becomes more and more important. Parts of the
gypsum model, at least the preparation, had to be digitized with an extraoral scanner to
create a digital threedimensional model. On this digital model, the restoration is desig-
ned on a computer with special design software and then milled out of a material block
in a CNC milling machine (Luthardt, Sandkuhl, Herold, & Walter, 2001). After milling, the
dental technician finishes the restoration at the gypsum model. The latest development
in CAD/CAM dentistry is the digital intraoral impression (DI)(Fasbinder, 2010)(Beuer,
Schweiger, & Edelhoff, 2008). The digital intraoral impression(DI) results in a threedi-
mensional virtual model from scanning the patients teeth directly inside the mouth.
Construction and milling can be carried out without a conventional intraoral impression
(CI) with the gypsum model and the following extraoral digitization. If needed, a physical
model can be by rapid prototyping (SLA, 3D-printing or miling) from the intraoral digital
impression data (Fasbinder, 2010).
A fundamental question, beside the clinical handling of the devices for the digital intra-
oral impression (DI) and the ease of the following steps in the digital workflow, is the ac-
curacy of this new impression technique.
Accuracy consists of precision and trueness (ISO 5725-1). Precision describes, how
close repeated measurements are to each other. The higher the precision, the more
predictable is the measurement. Trueness describes, how far the measurement is from
the real value of the measured object. A high trueness delivers a result that is close or
equal to the real value of the measured object.
Precision measurements compare the results of repeated measurements with a specific
impression method LIT)). Trueness measurement for conventional intraoral impression
with gypsum models are most frequently linear distance measurements (Chandran et
al., 2010)(Wostmann, Rehmann, & Balkenhol, 2009)(Hoyos & Soderholm, 2011). This
method is restricted to few measuring points, the need of specific geometric points with
clear markers for the measurement and the lack of displaying threedimensional changes
of the dental model like torsions and axis deviations (Chandran et al., 2010)(Brosky,
Pesun, Lowder, Delong, & Hodges, 2002)(Caputi & Varvara, 2008).
Threedimensional examination of trueness of impressions and gypsum models are rare
in literature. A demand is to know the real surface of the testing object. Therefore the
need of a reference scanner or the accessibility of a well known reference model to
measure the real surface of the testing object, e.g. the tooth or the dental arch, is the
limiting issue. It is possible to measure surfaces points with high trueness with CMM
machines. But these methods are lacking scan speed and the restriction of measuring
freeform surfaces like fissure lines and interproximal areas because of the geometrical
size of the tipball (Quaas, Rudolph, & Luthardt, 2007)(Delong, Heinzen, Hodges, Ko, &
Douglas, 2003). Optical scanners with high accuracy are currently limited to small mea-
surement fields like single teeth or quadrants (Luthardt, Kuhmstedt, & Walter,
2003)(Mehl, Ender, Mörmann, & Attin, 2009). Another method for accessing trueness of
a scanner is scanning calibrated objects like a sphere or material block of known size or
diameter (Vlaar & van der Zel, 2006)(Delong et al., 2003). These calibrated objects are
very small and do not have the typical morphology of teeth or the dental arch.
The aim of this study was
(1) to evaluate a new scanner with a specific scanning method to access high precise
and true surface scans of full arch models and
(2) verify, if the accuracy of this scanner is high enough to describe deviations from
conventional and digital impressions.
Materials and methods:
A metal model (Wirobond plus, Bego, Germany) of a patients upper dental arch was
made. Two full crown preparations (second premolar and second molar ind the second
quadrant) and one inlay preparation (first premolar in the first quadrant) were performed
(Image 1). This reference model was the base for all further conventional and digital im-
pressions.
1. Accuracy evaluation of the reference scanner, (a) precision and (b) trueness
(a) The reference model was scanned with a slightly reference scanner (InfiniteFocus
Standard, Alicona Imaging, Graz, Austria) five times in the same x-y-z direction with
the optimal scan settings for the model material. A 5x magnification objective was
used. The point size was 1,6x1,6µm in X-Y direction and 0.25µm in Z-direction. For
the most accurate scanning results, a horizontal base around the teeth is necessary
(Image1), that is scanned together with the tooth surface. Scan time for a full arch
model was 18 to 29 hours. After scanning, the base is trimmed with the „3D editor“
software module inside the IFM Software. The trimmed scan results in a data set of
about 20 million surface points (image 2). The five data sets were superimposed with
the difference analysis software (IFM software 3.5.0.1, Alicona Imaging, Graz, Aus-
tria) with a best fit algorithm. The software computes the difference from each mea-
suring point of one scan to the next neighbour point of the second scan. A difference
image of the two superimposed scans is created (image 3) and saved as an image
for visual analysis of deviations. The signed nearest neighbour distances of each
surface point between scan 1 and scan 2 are saved. This file is exported to Microsoft
Excel (Excel:mac 2011, Microsoft, Unterschleissheim, Germany) and the 90-10%/2
percentile is computed. This value represents the range where 80 percent of all dif-
ferences between the two superimposed scans are placed. The average of all 90-
10%/2 percentiles is taken as the result for precision of the reference scanner and
displayed in table 1.
(b) The reference model was scanned additionally four times in a rotated position of
about 90 degress around the z axis and 10 to 20 degrees in x-y axis. These scans
were compared to the first non rotated scan in the same manner. The results of the-
se comparisons are displayed in table 2. The rotated scans reveal any filter effects,
calibrating errors and matching errors of the software and are the validation of the
trueness of the scanner.
2. Accuracy of conventional impression material and gypsum model fabrication
Five conventional impressions were taken with a vinysiloxanether impression material
(Identium, Kettenbach, Eschenburg, Germany) and metal stock trays (ASA Perma-
Lock, ASA Dental, Bozzano, Italien) in a putty and wash technipue. The impression ma-
terial was set for 10 min on the model and then the impression removed from front to
back of the model. After eight hours the impression was poured with type IV gypsum
(Cam-Base, Dentona AG, Dortmund, Deutschland ) and set in upright position to
harden. After 40 min, the models were removed from the impression and stored for 24
to 48 hours at room temperature before scanning with the reference scanner. The mo-
dels were scanned with the reference scanner (IFM 3.5.0.1, Alicona Imaging, Graz,
Austria) with the occlusal surface oriented horizontal at the centre of the y-x table and
again with the horizontal base around the dental arch. The measure point size was
1.6µmx1.6µm in x-y direction and 0.25µm z-direction. After trimming the horizontal
base, each scan resulted in a three-dimensional model with about 20 million measured
surface points.
(a) The model scans were compared to each other to aquire the value for precision of
the conventional impression and
(b) compared to the scan of the reference model to determine the trueness of the con-
ventional impression method.
The difference analysis was performed in the same way as described in number 1(a).
3. Accuracy of digital impressions
Five digital impressions of the reference model were taken with the CEREC AC System
(Sirona Dental Company, Bensheim, Germany) using the CEREC Connect Software
3.82. The reference model was matted with OptiSpray (Sirona Dental Systems, Bens-
heim, Germany) and scanned with around 20 optical impressions to aquire the entire
dental arch. The resulting model was exported tas an STL file and imported into the
Alicona IFM Software for comparison to the reference model scanned with the Alicona
IFM device.
(a) The digital impressions were superimposed to each other and the differences display
the precision of the digital impression.
(b) The superimposition of the digital impressions with the reference model gives the
basisfor the trueness of the digital impression method.
The difference analysis was performed in the same way as described in number 1(a).
The independent sample t-test was used to analyse statistical differences between the
groups 1(a), 2(a) and 3(a) for the precision as well as 1(b), 2(b) and 3(b) for trueness of
the different impression methods with p<0.05 (IBM SPSS Statistics Version 19, IBM
SPSS, Chicag, USA).
Results
1. Accuracy evaluation of the reference scanner, (a) precision and (b) trueness
Image 4 shows the result of the precision measurement of the reference scanner. The
90-10% /2 quantil of all comparisons are located between ±0.5 and ± 2.5 with a mean
of 1.6±0.6 µm (median 2.0 µm) over the entire dental arch.
Image 4: boxplot of precision measurement
The difference images (image 5 a,b) show the coloured differences between two repea-
ted scans in a range from -10µm to +10µm. A homogenous deviation over the entire
dental arch with maximum values of 5µm on steep oral inclines of the incisor and the
second molar is visible. The matching areas of the single images are visible in these
special regions and show a higher deviatin than less angulated surfaces like the oc-
clusal area of molars and incisal regions of the anteriors. This error occur on almost
every scan and is addicted to the scanning process and image stitching routines. The
error does not cumulate over larger areas but is only in these particular regi-
ons.
Image 5(a,b): Difference images of two repeated scans of the master model. The difference color map is
set from -10µm to +10µm. A homogenous difference mapping of ±2µm is visible with local peaks to ±5µm
Image 6 shows the results of the trueness measurement of the reference scanner. The
90-10%/2 Quantil of all comparisons are located between ±3.5 to ±6.5 µm with a mean
of 5.3±1.1 µm (median 5.5 µm) over the entire dental arch.
Image 6: boxplot of trueness measurement
The difference images show the highest deviation in the second molar region of the first
quadrant and on the oral flanks of the canine (Image 6a,b). Maximum deviations are
about 8µm again in steep flanks of the tooth surface. The first quadrant shows a little
compression of the first and second molar and the second quadrant reveals a slight ver-
tical rotation of the dental arch from negativ deviations of the oral walls in the anterior to
positiv deviations of the oral walls in the posterior region.
Image 7(a,b): Difference images of two rotated scans. The difference color map is set from -10µm to
+10µm. Large areas differ up to 4µm while the steep flanks of the anteriors and second molars show so-
me higher differences up to 10µm.
2. Accuracy of conventional impression and gypsum model
The test group of conventional impressions with gypsum model revealed a mean preci-
sion of 12.5±2.5 µm (median 11.0µm) (image 4). The mean deviation from the reference
model is 20.4±2.2 µm (median 21.5µm) (trueness of the gypsum model) (image 6). The
values of the single comparisons are close to another, showing a high reliability of the
conventional impression in this in-vitro setup. The independent sample T-Test shows a
highest significant difference to the values from the reference scanner (p<0.001).
The visual evaluation of the precision measurement show small deviations in the anteri-
or and premolar regions of around 10µm and the higher and irregular occuring discre-
pancies of the second molar with maximum values up to 50µm (Image 7
a,b).
Image 8 a,b: Difference map of two repeated scans. The color difference map is set from -50µm to
+50µm. While in the anterior region only little deviations up to 20µm occur, the distal end of the arch
shows irregular deformations up to 50µm.
The visual examination of the difference images from the trueness show very low devia-
tions in the anterior region from canine to canine. In the premolar and molar region, the
deviations are higher and the gypsum model is slightly larger than the reference model.
At the distal end of the dental arch, irregular deviations occur with the highest deviations
of up to 50µm.
Image 9 a,b: Difference image of trueness measurement. The difference color map is set from -50 to
+50µm. 9a shows that the gypsum model is slightly larger than the reference model with rising deviation
to the distal end of the dental arch. 9b shows additional a distortion in the region of the second molar of
the second quadrant.
3. Digital impression
The digital impression method with the CEREC Bluecam showed a precision of
32.4±9.6 µm (median 31.7µm). The differences of virtual from the reference model
(trueness) are 58.6±15.8 µm (median 50µm). The independent sample T-Test shows
highest significant differences to the values from the reference scanner and conventio-
nal impression group (p<0.001). The difference images (Image 10,11) show the colou-
red differences between -50µm and +50µm.
The differences of the precision images show irregular deviation pattern. The anterior
region is more precise than the posterior and the higher posterior deviations are located
only at one side of the model (image 10 a,b).
Image 10 a,b: Difference image of two repeated digital impressions. The color difference map is set from -
50µm to +50 µm. Irregular deviations are visible across the entire dental arch. The highest differences are
located at the distal end. A wavelike distortion of the dental arch with alternating positive and negative
deviations together with a rotation of the anteriors is visible.
The visual analysis of the trueness shows a systematic deviation of the virtual 3D mo-
dels to the reference model with negativ values in the anterior and molar region and po-
sitive value in the canine and premolar region (image 11 a,b). Maximum differences in
the second molar area occur up to 170µm. The model is distorted along the sagittal axis
on both sides (image 11a,b).
Image 11 a,b: Difference map of trueness measurement of the digital impression. Color difference map is set to -
50µm to +50µm. A systematic distortion along the sagittal axis is visible. Negative deviations are located in the ante-
rior and molar region and positive deviations in the premolar regions up to 50µm. The distal end of the dental arch
differs up to 170µm
Discussion
Basis of our testing method is a scanner based on focus variation. A modulated white
light is projected onto the model surface. The reflected light is measured with an objecti-
ve with a low field of depth. The system is capable to measure large objects up to
10x10x10cm edge length. For small objects, different magnification lenses can be used
to reduce the measure point size down to 50nm.
Evaluation of trueness and precision multiple measurements of a testing object from dif-
ferent directions and angulations were compared. Filter algorithms and calibration errors
of the scanner will be visible by comparison of scans from different directions. However,
not visible in this testing procedure is only a linear scale error in x-, y- and z- axis with
the same amount. But this would be easily detectable with a length measure of a testing
normal.
According to our results, trueness and precision of the new reference scanner is extre-
mely high for scanning dental morphologies of a full arch model. For comparison, the
Laserscan 3D pro, serving as reference before yielded a trueness of 10-13µm scanning
a quadrant (Mehl, Gloger, Kunzelmann, & Hickel, 1997)(Mehl et al., 2009). Up to now,
we are not aware of any other system, that is possible to scan sophisticated tooth
surfaces with this high trueness and precision over an area up to 6cm in square. Other
studies work with geometrical forms, that are verified with CMM machines. CMM`s also
show high trueness and precision. But these machines are only aquiring a small amount
of points from the models surface. And for a precise model with CMM, you need to know
the shape of the surface before scanning (del corso 2009lit). That is ,for individual tooth
surfaces, not possible. Also the tip of the tactile probe has a certain diameter. That me-
ans, little morphological structures like fissure lines and gingival margins cannot be de-
tected with these systems. With the new reference scanner it is possible to aquire the
dental surface without prior knowledge of the morphology. Some manufacturer gave
their dental scanners accuracies from 10 µm to 30 µm. But these values are measured
on small models or geometrical defined testing blocks with up to 10mm length (Quelle
van der Zel).
In many studies, models are scanned with reference scanners, where the accuracy
ofthe scanner is not described (lit). Persson et al (lit) used the same digitization device
for reference and testing measurements. So the accuracy of the reference scanner is
not higher than the deviations they measure. The new reference scanner delivers an
accuracy that is significant (p<0.001) higher than the deviations from the conventional
and digital impression. The comparison of only 80 percent of the scanned surface is ad-
dicted to the measurement process. Neither there is a manual filtering of outliers or data
errors nor manual trimming the model to specific geometries. So we do have certain
scattered surface points with errors and areas, with only little surface points. The margin
of every scanned model is also slightly different. In these areas, the difference measu-
rement is not predictable to find the correct nearest signed neighbour. Therefore a
certain area, the lowest 10% and the highest 10% of the difference values is not taken
for comparisons. Yet with 80 percent of the scanned surface, the difference analysis re-
presents more model surface than a mean value with root mean square deviation that
only consists of 66% of the measured differences, a value which mainly was used in
other studies (Luthardt et al., 2003)(Quaas et al., 2007). Other studies use the accuracy
of the final restoration for evaluation the impression technique (Syrek et al.,
2010)(Del'Acqua, Arioli-Filho, Compagnoni, & Mollo Fde, 2008). This measurements are
in the end again only linear distance measurements and cannot evaluate threedimensi-
onal deviations of the impression method.
With a accuracy between 3µm and 10µm, the reference scanner is significant below the
deviations of conventional full arch impressions and therefore suitable for these ac-
curacy and precision measurements. With the increasing possibilities of CAD/CAM sys-
tems, like virtual articulation, production of removable partial or complete dentures, it is
very important to gain accurate data not only from the preparation, but also from the
entire dental arch.
The deviation of the gypsum model from the conventional impression with a polyvi-
nyether material show very low deviation across the entire dental arch. The higher de-
formations in the second molar regions may be due to the standard impression tray.
Brosky et al also found impressions with larger deformations in a testing group with
standard impression trays, however not located in the same area (Brosky, Major, De-
long, & Hodges, 2003). In this study he measured mean deviations from 27µm to
312µm were most measurements are ranged from 27 to 83µm. This result is, however a
different analysis method comparing more percent of the scanned surface was used, in
good company to our results for the conventional impression trueness. The differences
of this threedomensional difference analysis can`t be compared to studies with linear
distance measurements with accuracies of about 10µm for conventional impression ma-
terials (Wostmann et al., 2009)(Balkenhol, Ferger, & Wostmann, 2007)(Caputi & Varva-
ra, 2008)(Chandran et al., 2010).
The digital impressions with the CEREC Bluecam show higher deviations from the refe-
rence model resulting in a significant (p<0.001) lower precision and trueness compared
to the conventional impression group. The pattern of deviation in the sagittal axis in one
quadrant was also revealed in former studies (Mehl et al., 2009) (Luthardt, Loos, &
Quaas, 2005). The deviations in the anterior region may be due to the creation of the full
arch scan by combining two overlapping partial scans of both quadrants (Manufacturers
recommendation, Cerec Connect Help system, Operators manual, Full Arch Impressi-
on). The registration area is the anterior region from canine to canine. In the anterior
regions, with less structured tooth surface and steep inclines, more error with the optical
impression can occur. It is most assumable, that the superimposition process leads to
that kind of deviations. These two errors seems to be systematic and maybe reduced or
passed by with further software improvement.
Balkenhol, M., Ferger, P., & Wostmann, B. (2007). Dimensional accuracy of 2-stage
putty-wash impressions: influence of impression trays and viscosity Int J Prosthodont,
20(6), 573-575. United States.
Beuer, F., Schweiger, J., & Edelhoff, D. (2008). Digital dentistry: an overview of recent
developments for CAD/CAM generated restorations Br Dent J, 204(9), 505-511.
doi:10.1038/sj.bdj.2008.350
Brosky, M., Major, R., Delong, R., & Hodges, J. (2003). Evaluation of dental arch repro-
duction using three-dimensional optical digitization J Prosthet Dent, 90(5), 434-440.
Brosky, M., Pesun, I., Lowder, P., Delong, R., & Hodges, J. (2002). Laser digitization of
casts to determine the effect of tray selection and cast formation technique on accuracy
J Prosthet Dent, 87(2), 204-209.
Caputi, S., & Varvara, G. (2008). Dimensional accuracy of resultant casts made by a
monophase, one-step and two-step, and a novel two-step putty/light-body impression
technique: an in vitro study J Prosthet Dent, 99(4), 274-281.
Chandran, D., Jagger, D., Jagger, R., & Barbour, M. (2010). Two- and three-
dimensional accuracy of dental impression materials: effects of storage time and mo-
isture contamination Biomed Mater Eng, 20(5), 243-249. Netherlands.
doi:10.3233/BME-2010-0638
Del'Acqua, M., Arioli-Filho, J., Compagnoni, M., & Mollo Fde, A. J. (2008). Accuracy of
impression and pouring techniques for an implant-supported prosthesis Int J Oral Maxil-
lofac Implants, 23(2), 226-236.
Delong, R., Heinzen, M., Hodges, J., Ko, C., & Douglas, W. (2003). Accuracy of a sys-
tem for creating 3D computer models of dental arches J Dent Res, 82(6), 438-442.
Fasbinder, D. (2010). Digital dentistry: innovation for restorative treatment Compend
Contin Educ Dent, 31 Spec No 4, 2-11; quiz 12. United States.
Hoyos, A., & Soderholm, K. (2011). Influence of tray rigidity and impression technique
on accuracy of polyvinyl siloxane impressions Int J Prosthodont, 24(1), 49-54. United
States.
Luthardt, R., Kuhmstedt, P., & Walter, M. (2003). A new method for the computer-aided
evaluation of three-dimensional changes in gypsum materials Dent Mater, 19(1), 19-24.
Luthardt, R., Loos, R., & Quaas, S. (2005). Accuracy of intraoral data acquisition in
comparison to the conventional impression International journal of computerized den-
tistry, 8(4), 283-294.
Luthardt, R., Sandkuhl, O., Herold, V., & Walter, M. (2001). Accuracy of mechanical digi-
tizing with a CAD/CAM system for fixed restorations Int J Prosthodont, 14(2), 146-151.
Mehl, A., Ender, A., Mörmann, W., & Attin, T. (2009). Accuracy testing of a new intraoral
3D camera International journal of computerized dentistry, 12(1), 11-28.
Mehl, A., Gloger, W., Kunzelmann, K., & Hickel, R. (1997). A new optical 3-D device for
the detection of wear J Dent Res, 76(11), 1799-1807.
Perakis, N., Belser, U., & Magne, P. (2004). Final impressions: a review of material pro-
perties and description of a current technique Int J Periodontics Restorative Dent, 24(2),
109-117.
Persson, A., ODEN, A., ANDERSSON, M., & Sandborgh-Englund, G. (2009). Digitizati-
on of simulated clinical dental impressions: virtual three-dimensional analysis of
exactness Dent Mater, 25(7), 929-936.
Piwowarczyk, A., Ottl, P., Buchler, A., Lauer, H., & Hoffmann, A. (2002). In vitro study
on the dimensional accuracy of selected materials for monophase elastic impression
making Int J Prosthodont, 15(2), 168-174.
Quaas, S., Rudolph, H., & Luthardt, R. (2007). Direct mechanical data acquisition of
dental impressions for the manufacturing of CAD/CAM restorations J Dent, 35(12), 903-
908.
Syrek, A., Reich, G., Ranftl, D., Klein, C., Cerny, B., & Brodesser, J. (2010). Clinical
evaluation of all-ceramic crowns fabricated from intraoral digital impressions based on
the principle of active wavefront sampling J Dent, 38(7), 553-559. England.
doi:10.1016/j.jdent.2010.03.015
Vlaar, S., & van der Zel, J. (2006). Accuracy of dental digitizers Int Dent J, 56(5), 301-
309.
Wettstein, F., Sailer, I., Roos, M., & Hammerle, C. (2008). Clinical study of the internal
gaps of zirconia and metal frameworks for fixed partial dentures Eur J Oral Sci, 116(3),
272-279. Denmark. doi:10.1111/j.1600-0722.2008.00527.x
Wostmann, B., Rehmann, P., & Balkenhol, M. (2009). Accuracy of impressions obtained
with dual-arch trays Int J Prosthodont, 22(2), 158-160.